Instant Personalized Large Language Model Adaptation via Hypernetwork

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Tan, Zhaoxuan, Zhang, Zixuan, Wen, Haoyang, Li, Zheng, Zhang, Rongzhi, Chen, Pei, Mo, Fengran, Liu, Zheyuan, Zeng, Qingkai, Yin, Qingyu, Jiang, Meng
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908600634441728
author Tan, Zhaoxuan
Zhang, Zixuan
Wen, Haoyang
Li, Zheng
Zhang, Rongzhi
Chen, Pei
Mo, Fengran
Liu, Zheyuan
Zeng, Qingkai
Yin, Qingyu
Jiang, Meng
author_facet Tan, Zhaoxuan
Zhang, Zixuan
Wen, Haoyang
Li, Zheng
Zhang, Rongzhi
Chen, Pei
Mo, Fengran
Liu, Zheyuan
Zeng, Qingkai
Yin, Qingyu
Jiang, Meng
contents Personalized large language models (LLMs) tailor content to individual preferences using user profiles or histories. However, existing parameter-efficient fine-tuning (PEFT) methods, such as the ``One-PEFT-Per-User'' (OPPU) paradigm, require training a separate adapter for each user, making them computationally expensive and impractical for real-time updates. We introduce Profile-to-PEFT, a scalable framework that employs a hypernetwork, trained end-to-end, to map a user's encoded profile directly to a full set of adapter parameters (e.g., LoRA), eliminating per-user training at deployment. This design enables instant adaptation, generalization to unseen users, and privacy-preserving local deployment. Experimental results demonstrate that our method outperforms both prompt-based personalization and OPPU while using substantially fewer computational resources at deployment. The framework exhibits strong generalization to out-of-distribution users and maintains robustness across varying user activity levels and different embedding backbones. The proposed Profile-to-PEFT framework enables efficient, scalable, and adaptive LLM personalization suitable for large-scale applications.
format Preprint
id arxiv_https___arxiv_org_abs_2510_16282
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Instant Personalized Large Language Model Adaptation via Hypernetwork
Tan, Zhaoxuan
Zhang, Zixuan
Wen, Haoyang
Li, Zheng
Zhang, Rongzhi
Chen, Pei
Mo, Fengran
Liu, Zheyuan
Zeng, Qingkai
Yin, Qingyu
Jiang, Meng
Computation and Language
Personalized large language models (LLMs) tailor content to individual preferences using user profiles or histories. However, existing parameter-efficient fine-tuning (PEFT) methods, such as the ``One-PEFT-Per-User'' (OPPU) paradigm, require training a separate adapter for each user, making them computationally expensive and impractical for real-time updates. We introduce Profile-to-PEFT, a scalable framework that employs a hypernetwork, trained end-to-end, to map a user's encoded profile directly to a full set of adapter parameters (e.g., LoRA), eliminating per-user training at deployment. This design enables instant adaptation, generalization to unseen users, and privacy-preserving local deployment. Experimental results demonstrate that our method outperforms both prompt-based personalization and OPPU while using substantially fewer computational resources at deployment. The framework exhibits strong generalization to out-of-distribution users and maintains robustness across varying user activity levels and different embedding backbones. The proposed Profile-to-PEFT framework enables efficient, scalable, and adaptive LLM personalization suitable for large-scale applications.
title Instant Personalized Large Language Model Adaptation via Hypernetwork
topic Computation and Language
url https://arxiv.org/abs/2510.16282